CN106789721B - Intelligent QOS method and system based on token bucket - Google Patents
Intelligent QOS method and system based on token bucket Download PDFInfo
- Publication number
- CN106789721B CN106789721B CN201710184142.1A CN201710184142A CN106789721B CN 106789721 B CN106789721 B CN 106789721B CN 201710184142 A CN201710184142 A CN 201710184142A CN 106789721 B CN106789721 B CN 106789721B
- Authority
- CN
- China
- Prior art keywords
- queue
- data packet
- user
- token bucket
- priority
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/215—Flow control; Congestion control using token-bucket
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/76—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/80—Actions related to the user profile or the type of traffic
- H04L47/805—QOS or priority aware
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides an intelligent QOS method and system based on a token bucket, and belongs to the technical field of network application. The method comprises the following steps: marking the message priority, and then sending the data packet marked with the priority into a user security token bucket; judging whether the bandwidth is in the guaranteed bandwidth, if so, sending the data packet into a system guarantee queue, and if not, sending the data packet into a user queue according to user classification; the scheduler preferentially calls the data packets of the system guarantee queue into a system token bucket according to the priority order, and if the residual bandwidth exists, calls the data packets of the user queue into the system token bucket; the scheduler sends the data packet according to the token of the system token bucket. The invention has the beneficial effects that: on the principle of keeping fairness, the system adds a punishment mechanism to ensure that part of applications do not occupy a large amount of bandwidth any more, the bandwidth is distributed to other applications, the delay is obviously reduced, and the internet surfing environment of a user is obviously improved.
Description
Technical Field
The invention relates to the technical field of network application, in particular to an intelligent QOS method based on a token bucket and a system for realizing the intelligent QOS method.
Background
With the development of the IP technology and the network, all messages in the network are treated indiscriminately in the past, and the uncertain and unreliable transmission modes such as first-in first-out and best-effort message delivery are adopted, so that the requirements of various complex network services emerging in the current network, such as various applications, video, voice, and the like, are no longer met.
Most of the conventional QOS is based on flow control of queuing and congestion processing based on one flow (data flow identified by five-tuple), and different application services are not distinguished, so that the requirement of reducing the service quality of other applications for guaranteeing the bandwidth of some applications cannot be realized under the condition that the total network resources are limited.
In the daily office process of an enterprise, different employees have different requirements on network application, irrelevant services usually occupy a large amount of bandwidth resources, when a network is overloaded or congested, a data packet of a core service is delayed or discarded because of no corresponding bandwidth guarantee, and because the total bandwidth is fixed, a set of scheme for reasonably allocating the bandwidth resources of each user needs to be provided to solve the problem of bandwidth resource preemption caused by service requirements so as to achieve the purposes of effectively managing internet traffic and improving network service quality, and dynamically and reasonably allocating the bandwidth.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides an intelligent QOS method based on a token bucket and a system for realizing the intelligent QOS method.
The intelligent QOS method based on the token bucket comprises the following steps:
s1: marking the message priority, and then sending the data packet marked with the priority into a user security token bucket;
s2: judging whether the bandwidth is in the guaranteed bandwidth, if so, sending the data packet into a system guarantee queue, and if not, sending the data packet into a user queue according to user classification;
s3: the scheduler preferentially calls the data packets of the system guarantee queue into a system token bucket according to the priority order, and if the residual bandwidth exists, calls the data packets of the user queue into the system token bucket;
s4: the scheduler sends the data packet according to the token of the system token bucket.
The invention further improves, after the step S3 and before the step S4 is executed, the method further comprises the steps of: and mounting the data packet on a timer linked list, and triggering and calling the data packet into a system token bucket by the timer.
The invention is further improved, when the data packet enters and exits the user guarantee token bucket and the system token bucket, the timer triggers to increase the tokens at regular intervals, and the consumed tokens are updated in real time when the data packet enters.
The present invention is further improved in that in step S2, the system guarantee queue is shared by all users, and the data packets of the users with the rate within the guaranteed bandwidth directly enter the system guarantee queue to be sent according to the priority, wherein the value of the guaranteed bandwidth is equal to the total bandwidth divided by the number of online users.
The invention is further improved, the number of the sub-queues in the user queue is equal to the number of the online users, wherein the data packets in each sub-queue are arranged according to the priority order.
The invention is further improved, the processing method of the data packet entering the system guarantee queue or the user queue is as follows:
s21: carrying out enqueue processing;
s22: acquiring a priority of a data packet;
s23: judging whether the queue is full, if so, losing the packet and quitting; if not, entering the queue according to the priority and exiting.
In a further improvement of the present invention, in step S3, when the scheduler retrieves the packets from the user queue, the packets between the sub-queues are dequeued fairly.
The invention is further improved, in step S3, the system guarantee queue packet sending processing method is:
s301: the system guarantees queue dequeuing processing;
s302: the data packets of the system guarantee queue are sequentially output from high to low according to the priority order;
s303: judging whether a data packet exists, if not, executing step S305, and if so, executing step S304;
s304: judging wan whether the port broadband token is enough, if yes, sending a data packet, counting the flow, then executing step S302, if not, re-enqueuing the data packet, and marking wan that the password card is not enough;
s305: the system guarantee team lists the package completion.
The invention is further improved, and the user queue packet sending processing method comprises the following steps:
s311: starting to send a bag;
s312: the user queue data packets are output from high to low according to the priority;
s313: judging whether the package is successfully produced, if the package is successfully produced, executing the step S314, and if the package is unsuccessfully produced, executing the step S316;
s314: judging whether the user has a speed limit rule, if not, executing the step S315; if yes, judging whether the speed limit token is enough, and if so, executing the step S315;
s315: judging wan whether the password card is enough, if yes, sending the data packet, counting the flow, then executing step S312, if no, re-enqueuing the data packet, and marking wan that the password card is not enough;
s316: and exiting the hair-package.
The invention also provides a system for realizing the intelligent QOS method, which comprises a message identification module: the message type identification module is used for identifying the type of the message; a message marking module: the priority used for marking the message; a QOS kernel implementation module: the system comprises a user queue, a system guarantee queue and a user queue, wherein the user queue is used for judging whether the bandwidth is in the guarantee bandwidth or not, if so, the data packet is sent to the system guarantee queue, and if not, the data packet is sent to the user queue according to user classification; wherein, the QOS kernel implementation module comprises: a scheduler: the system token bucket is used for preferentially calling the data packets of the system guarantee queue into the system token bucket according to the priority order, calling the data packets of the user queue into the system token bucket if the residual bandwidth exists, and sending the data packets according to the tokens of the system token bucket; a token computation module: the method is used for calculating the number of user guarantee token buckets and system token buckets when data packets enter and exit; the intelligent QOS system further comprises an interface layer: the kernel module is used for realizing communication between the QOS kernel module and the application layer; the packet sending control interface module: and the system is used for sending the data packet to a specified address according to the uplink and downlink marks of the data packet.
Compared with the prior art, the invention has the beneficial effects that: the bandwidth is dynamically and reasonably distributed, when a plurality of devices all have bandwidth requirements, the problems that due to the fact that a lot of bandwidth is occupied by multiple connected applications such as P2P and the like, other users cannot surf the internet, games, normally watch videos and the like are solved, a punishment mechanism is added to the system on the principle that fairness is kept, parts of the applications and the like do not occupy a lot of bandwidth any more, the bandwidth is distributed to other applications, delay is obviously reduced, and the internet surfing environment of the users is obviously improved.
Drawings
FIG. 1 is a flow chart of the method of the present invention;
FIG. 2 is a schematic process flow diagram of the present invention;
FIG. 3 is a flow chart of packet queuing;
FIG. 4 is a flow chart of the process of dequeuing a user queue packet into a system guarantee queue;
FIG. 5 is a flow chart of a system guarantee queue packet processing method;
FIG. 6 is a flow chart of a user queue packet processing method;
fig. 7 is a block diagram of a system configuration.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
As shown in fig. 1 and 2, the intelligent QOS method based on token bucket of the present invention includes the following steps:
s1: marking the message priority, and then sending the data packet marked with the priority into a user security token bucket;
s2: judging whether the bandwidth is in the guaranteed bandwidth, if so, sending the data packet into a system guarantee queue, and if not, sending the data packet into a user queue according to user classification;
s3: the scheduler preferentially calls the data packets of the system guarantee queue into a system token bucket according to the priority order, and if the residual bandwidth exists, calls the data packets of the user queue into the system token bucket;
s4: the scheduler sends the data packet according to the token of the system token bucket.
The traditional FIFO (first in first out) queue does not classify the messages, and the messages are sent in a first in first out mode, so that the requirement of application priority of people cannot be met; although the SPQ queue (strict priority queue) adopts the concept of priority, when the high priority queue always has a packet, the low priority queue will starve; the CQ queue (customizable queue) technique improves on the SPQ technique to avoid the "starvation" phenomenon, but does not guarantee the delay problem for any flow.
Based on the above reasons, we improve and add a guarantee queue mechanism on the basis of the SPQ queue, design a guarantee queue shared by all users, and at the same time, each user has its own independent priority queue to store data packets, and no matter whether the guarantee queue or the user's own queue still uses an absolute priority packet-sending mechanism, only a high priority has no data packet to send a low priority data packet, this example divides the priority of a priority queue into 5 (0-4), and 4 is the lowest priority, and since the packet of the guarantee queue is sent first and then the packet sequence of the user's own queue is sent each time, it can be guaranteed that the flow of a new user is not delayed.
In order to better ensure the reasonable distribution of the bandwidth among the users, the embodiment sets a mechanism for ensuring the bandwidth, namely the minimum guaranteed rate of the users, and the value of the ensured bandwidth is equal to the total bandwidth divided by the number of online users; the data packets of the users in the guaranteed bandwidth at the speed directly enter the priority of the system guarantee queue to be sent, and the users who just go online can quickly seize the guaranteed bandwidth from other users.
Further, on the basis of the above method, the present invention preferably further comprises, after step S3 and before step S4 is executed, a timer triggering step of: and mounting the data packet on a timer linked list, and triggering and calling the data packet into a system token bucket by the timer.
The packet sending method is characterized in that packet sending is realized by combining data packet triggering and timer triggering, each data packet is triggered to send a system guarantee queue data packet preferentially after entering the system, and then the data packet of the user queue is sent; if the user queue has data packet, the data packet is mounted on a timer linked list and triggered to be sent by the timer, the timer traverses the linked list, the guarantee queue is sent first, wan port bandwidth is remained, and the data packet on the user queue is sent continuously.
In addition, the traditional token bucket is optimized, specifically, for the traditional token bucket speed limit, each data packet triggers the calculation and updating of time and token number, and one data packet can be calculated for multiple times to consume performance; in the embodiment, the increase and the consumption of the token are processed separately, and the increase of the token is triggered by the timer, for example, the token is increased once in 0.01s (the number of the token is increased each time, and the number of bytes which can pass through 0.01s is calculated according to the limited rate), the data packet enters the calculation which only needs to consume the token, the consumed token is the length of the data packet, a large amount of calculation is reduced, and the performance is improved.
The specific processing method of the embodiment is as follows:
(1) and the data packet enters the intelligent QOS system for processing, and is directly exited without being started. If the user priority queue is started, judging whether a data packet still exists in the user priority queue;
(2) and if the user queue has no data packet, the user enters the user priority queue if the user exceeds the guaranteed rate, enters the system guaranteed queue if the user does not exceed the guaranteed rate, consumes the token if the user successfully enters the system guaranteed queue, and updates the token of the speed limit rule for the user with the speed limit rule. As shown in fig. 3, in step S2, the specific method for entering the data packet into the user queue or the system guarantee queue includes:
s21: carrying out enqueue processing;
s22: acquiring a priority of a data packet;
s23: judging whether the queue is full, if so, losing the packet and quitting; if not, entering the queue according to the priority and exiting.
(3) As shown in fig. 4, if there is a data packet in the user queue, the user queue is entered, the data packet in the user queue is circularly dequeued, and if there is no packet or the guaranteed rate is exceeded, the circulation is exited. The specific dequeuing processing method for the user queue comprises the following steps:
a: the user queue dequeuing processing is carried out, and a speed limit sign and a residual token are obtained based on a user;
b: each sub-queue sequentially outputs packets from high to low according to the priority order, and if no data packet exists, the process is finished directly;
c: and judging whether the token of the system token bucket is enough, if not, directly ending, if so, updating the rest tokens of the rule speed limit when the data packet has the rule and the guaranteed bandwidth is greater than the speed limit, and dequeuing the data packet.
(4) As shown in fig. 5, in step S3, the data packet in the system guarantee queue is sent until it is empty or the total bandwidth is exceeded, and the system guarantee queue exits from sending, where the system guarantee queue sends the data packet by the following method:
s301: the system guarantees queue dequeuing processing;
s302: the data packets of the system guarantee queue are sequentially output from high to low according to the priority order;
s303: judging whether a data packet exists, if not, executing step S305, and if so, executing step S304;
s304: judging wan whether the port broadband token is enough, if yes, sending a data packet, counting the flow, then executing step S302, if not, re-enqueuing the data packet, and marking wan that the password card is not enough;
s305: the system guarantee team lists the package completion.
(5) As shown in fig. 6, if the total bandwidth is still left, the data packet of the user queue is continuously transmitted, and when the total bandwidth is exceeded or the speed limit of the user is exceeded (if the speed limit rule is exceeded), the user packet sending processing is exited, and the user queue packet sending processing method includes:
s311: starting to send a bag;
s312: the user queue data packets are output from high to low according to the priority;
s313: judging whether the package is successfully produced, if the package is successfully produced, executing the step S314, and if the package is unsuccessfully produced, executing the step S316;
s314: judging whether the user has a speed limit rule, if not, executing the step S315; if yes, judging whether the speed limit token is enough, and if so, executing the step S315;
s315: judging wan whether the password card is enough, if yes, sending the data packet, counting the flow, then executing step S312, if no, re-enqueuing the data packet, and marking wan that the password card is not enough;
s316: and exiting the hair-package.
(6) In the packet sending process, when a user queue has data packets, the user is hung on a linked list of a timer, the timer triggers the packet, no residual data packets exist, and the user is removed from the linked list when the user is on the linked list of the timer.
(7) And triggering a timer, traversing the user, firstly sending a guarantee queue data packet, and continuously sending the data packet of the user queue if the bandwidth is remained.
As shown in fig. 7, the present invention further provides a system for implementing the intelligent QOS method, including a message identification module: the message type identification module is used for identifying the type of the message; a message marking module: the priority used for marking the message; a QOS kernel implementation module: the system comprises a user queue, a system guarantee queue and a user queue, wherein the user queue is used for judging whether the bandwidth is in the guarantee bandwidth or not, if so, the data packet of the message is sent to the system guarantee queue, and if not, the data packet is sent to the user queue according to the user classification; wherein, the QOS kernel implementation module comprises: a scheduler: the system token bucket is used for preferentially calling the data packets of the system guarantee queue into the system token bucket according to the priority order, calling the data packets of the user queue into the system token bucket if the residual bandwidth exists, and sending the data packets according to the tokens of the system token bucket; a token computation module: the method is used for calculating the number of user guarantee token buckets and system token buckets when data packets enter and exit; the intelligent QOS system further comprises an interface layer: the kernel module is used for realizing communication between the QOS kernel module and the application layer; the packet sending control interface module: and the system is used for sending the data packet to a specified address according to the uplink and downlink marks of the data packet.
The QOS kernel implementation module of the embodiment comprises two implementation modes, namely an intelligent speed limit mode and an intelligent priority mode, and meanwhile, only one of the two modes can be selected for configuration, and the WEB configuration page of the application layer realizes the mode configuration of the QOS kernel module through the interface layer. The following two modes are described separately:
(1) the intelligent speed limiting mode is as follows: the traditional speed limit function is only to configure CAR for a single user, while the intelligent speed limit of the embodiment fuses the traditional speed limit function and application priority, the default is to place the preemption resource at low priority (such as thunder), when the speed limit is performed for a single PC, the thunder can be suppressed, the data of other applications can be sent preferentially, the bandwidth is not completely occupied by the thunder, when no other applications are used, the thunder immediately uses all the bandwidth, thus, when the speed limit is performed, the bandwidth can be reasonably allocated to each application of the PC instead of being completely occupied by a certain application.
(2) Intelligent priority mode: four modes of video priority, webpage priority, game priority and download priority can be configured for each user. The priority of the application is set according to different requirements of users, so that the occupied bandwidth of the application required by the users is guaranteed.
As an embodiment of the present invention, the priority processing method in this embodiment is:
the application priorities correspond to priority queues, and different applications enter priority queues of different levels according to the priorities. The application priority of the embodiment is identified and set by the message identification module, and the application priority is marked on the message marking module, and the intelligent priority mode user can modify the application priority according to the requirement; and for the data stream which is not identified as the application, determining the priority according to the number of data stream packets:
1) ICMP (Internet Control Message Protocol), DNS (Domain Name System), and the top 5 packets of connection have the highest priority (1).
2) The first 50 packets of the HTTP protocol flow with highest priority (1), less than 250 packets with high priority (2), less than 350 packets with medium priority (3), more than 350 packets are considered as HTTP download-like applications, and with low priority (4).
3) TCP (Transmission Control Protocol) or UDP (User Datagram Protocol) connects the first 200 packets with medium priority (3), and changes to low priority (4) if more than 200 packets.
The above-described embodiments are intended to be illustrative, and not restrictive, of the invention, and all changes that come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Claims (9)
1. An intelligent QOS method based on token bucket is characterized by comprising the following steps:
s1: marking the message priority, and then sending the data packet marked with the priority into a user security token bucket;
s2: judging whether the bandwidth is in the guaranteed bandwidth, if so, sending the data packet into a system guarantee queue, and if not, sending the data packet into a user queue according to user classification;
s3: the scheduler preferentially calls the data packets of the system guarantee queue into a system token bucket according to the priority order, and if the residual bandwidth exists, calls the data packets of the user queue into the system token bucket;
s4: the scheduler sends packets based on the tokens of the system token bucket,
after step S3, before step S4 is executed, the method further includes the steps of: the data packet is mounted on a timer linked list, the timer triggers and calls the system token bucket,
the packet sending is the combination of data packet triggering and timer triggering, each data packet is triggered to send a system guarantee queue data packet preferentially after entering the system, and then the data packet of the user queue is sent; if the user queue has data packet, the data packet is mounted on a timer linked list and triggered to be sent by the timer, the timer traverses the linked list, the guarantee queue is sent first, wan port bandwidth is remained, and the data packet on the user queue is sent continuously.
2. The intelligent QOS method of claim 1, wherein: when the data packet enters and exits the user guarantee token bucket and the system token bucket, the timer triggers to increase tokens at regular intervals, and the consumed tokens are updated in real time when the data packet enters.
3. The intelligent QOS method of claim 1 or 2, wherein: in step S2, the system guarantee queue is shared by all users, and the data packets of the users with the rate within the guaranteed bandwidth directly enter the system guarantee queue and are sent according to the priority, where the value of the guaranteed bandwidth is equal to the total bandwidth divided by the number of online users.
4. The intelligent QOS method of claim 3, wherein: the number of the sub-queues in the user queue is equal to the number of online users, wherein the data packets in each sub-queue are arranged according to the priority order.
5. The intelligent QOS method of claim 4, wherein: the processing method for the data packet to enter the system guarantee queue or the user queue comprises the following steps:
s21: carrying out enqueue processing;
s22: acquiring a priority of a data packet;
s23: judging whether the queue is full, if so, losing the packet and quitting; if not, entering the queue according to the priority and exiting.
6. The intelligent QOS method of claim 1 or 2, wherein: when the scheduler retrieves the packets from the user queue in step S3, the packets between the sub-queues are dequeued in a fair round.
7. The intelligent QOS method of claim 6, wherein: in step S3, the system guarantee queue packet sending processing method includes:
s301: the system guarantees queue dequeuing processing;
s302: the data packets of the system guarantee queue are sequentially output from high to low according to the priority order;
s303: judging whether a data packet exists, if not, executing step S305, and if so, executing step S304;
s304: judging wan whether the port broadband token is enough, if yes, sending a data packet, counting the flow, then executing step S302, if not, re-enqueuing the data packet, and marking wan that the password card is not enough;
s305: the system guarantee team lists the package completion.
8. The intelligent QOS method of claim 6, wherein: the user queue packet sending processing method comprises the following steps:
s311: starting to send a bag;
s312: the user queue data packets are output from high to low according to the priority;
s313: judging whether the package is successfully produced, if the package is successfully produced, executing the step S314, and if the package is unsuccessfully produced, executing the step S316;
s314: judging whether the user has a speed limit rule, if not, executing the step S315; if yes, judging whether the speed limit token is enough, and if so, executing the step S315;
s315: judging wan whether the password card is enough, if yes, sending the data packet, counting the flow, then executing step S312, if no, re-enqueuing the data packet, and marking wan that the password card is not enough;
s316: and exiting the hair-package.
9. An intelligent QOS system implementing the intelligent QOS method of any one of claims 1 to 8, comprising: a message identification module: the message type identification module is used for identifying the type of the message;
a message marking module: the priority used for marking the message;
a QOS kernel implementation module: the system comprises a user queue, a system guarantee queue and a user queue, wherein the user queue is used for judging whether the bandwidth is in the guarantee bandwidth or not, if so, the data packet is sent to the system guarantee queue, and if not, the data packet is sent to the user queue according to user classification;
wherein, the QOS kernel implementation module comprises:
a scheduler: the system token bucket is used for preferentially calling the data packets of the system guarantee queue into the system token bucket according to the priority order, calling the data packets of the user queue into the system token bucket if the residual bandwidth exists, and sending the data packets according to the tokens of the system token bucket; a token computation module: the method is used for calculating the number of user guarantee token buckets and system token buckets when data packets enter and exit;
the intelligent QOS system further comprises an interface layer: the kernel module is used for realizing communication between the QOS kernel module and the application layer;
the packet sending control interface module: and the system is used for sending the data packet to a specified address according to the uplink and downlink marks of the data packet.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710184142.1A CN106789721B (en) | 2017-03-24 | 2017-03-24 | Intelligent QOS method and system based on token bucket |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710184142.1A CN106789721B (en) | 2017-03-24 | 2017-03-24 | Intelligent QOS method and system based on token bucket |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106789721A CN106789721A (en) | 2017-05-31 |
| CN106789721B true CN106789721B (en) | 2021-01-05 |
Family
ID=58967849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710184142.1A Active CN106789721B (en) | 2017-03-24 | 2017-03-24 | Intelligent QOS method and system based on token bucket |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106789721B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109391555B (en) * | 2017-08-08 | 2022-08-12 | 迈普通信技术股份有限公司 | Message scheduling method and device and communication equipment |
| CN107682282B (en) * | 2017-09-15 | 2021-04-06 | 北京外通电子技术公司 | Service quality method and network equipment for guaranteeing service bandwidth |
| CN108173780B (en) * | 2017-11-29 | 2021-06-04 | 东软集团股份有限公司 | Data processing method, data processing device, computer and storage medium |
| CN110601996B (en) * | 2019-09-06 | 2021-06-04 | 无锡江南计算技术研究所 | Looped network anti-starvation flow control method adopting token bottom-preserving distributed greedy algorithm |
| CN112367270B (en) * | 2020-10-30 | 2023-01-10 | 锐捷网络股份有限公司 | Method and equipment for sending message |
| CN112671832A (en) * | 2020-12-03 | 2021-04-16 | 中国科学院计算技术研究所 | Forwarding task scheduling method and system for guaranteeing hierarchical time delay in virtual switch |
| CN112799861B (en) * | 2021-01-29 | 2024-01-09 | 上海弘积信息科技有限公司 | Method for realizing flow rate limiting lock-free concurrency under multi-core architecture |
| CN112953848B (en) * | 2021-03-12 | 2023-05-02 | 西安电子科技大学 | Traffic supervision method, system and equipment based on strict priority |
| CN113328957B (en) * | 2021-05-18 | 2023-05-23 | 深圳Tcl新技术有限公司 | Flow control method and device and electronic equipment |
| CN113676416B (en) * | 2021-10-22 | 2021-12-28 | 浙江锐文科技有限公司 | Method for improving network service quality in high-speed network card/DPU |
| CN113938439B (en) * | 2021-10-25 | 2023-12-15 | 深圳市风云实业有限公司 | Port queue scheduling method based on timer and electronic equipment |
| CN114679421A (en) * | 2022-04-19 | 2022-06-28 | 深圳市吉祥腾达科技有限公司 | Queue scheduling method and system based on user data flow priority |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101179486A (en) * | 2006-11-10 | 2008-05-14 | 中兴通讯股份有限公司 | Computer network data packet forwarding CAR queue management method |
| CN101202701A (en) * | 2006-12-12 | 2008-06-18 | 中兴通讯股份有限公司 | Method for distributing band width of assemblage useable bit rate transaction in grouping network |
| CN101938403A (en) * | 2009-06-30 | 2011-01-05 | 中国电信股份有限公司 | Assurance method of multi-user and multi-service quality of service and service access control point |
| CN105681112A (en) * | 2016-03-26 | 2016-06-15 | 华为技术有限公司 | Method of realizing multi-level committed access rate control and related device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7627675B2 (en) * | 2003-05-01 | 2009-12-01 | Cisco Technology, Inc. | Methods and devices for regulating traffic on a network |
-
2017
- 2017-03-24 CN CN201710184142.1A patent/CN106789721B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101179486A (en) * | 2006-11-10 | 2008-05-14 | 中兴通讯股份有限公司 | Computer network data packet forwarding CAR queue management method |
| CN101202701A (en) * | 2006-12-12 | 2008-06-18 | 中兴通讯股份有限公司 | Method for distributing band width of assemblage useable bit rate transaction in grouping network |
| CN101938403A (en) * | 2009-06-30 | 2011-01-05 | 中国电信股份有限公司 | Assurance method of multi-user and multi-service quality of service and service access control point |
| CN105681112A (en) * | 2016-03-26 | 2016-06-15 | 华为技术有限公司 | Method of realizing multi-level committed access rate control and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106789721A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106789721B (en) | Intelligent QOS method and system based on token bucket | |
| US9565125B2 (en) | Multicast to unicast conversion technique | |
| US7525911B2 (en) | Congestion controller for Ethernet switch | |
| US8144588B1 (en) | Scalable resource management in distributed environment | |
| WO2010020988A2 (en) | System and methods for distributed quality of service enforcement | |
| EP3073680B1 (en) | Methods, queueing system, network element and network system for queueing and processing of packets | |
| CN109391555B (en) | Message scheduling method and device and communication equipment | |
| EP2761826A1 (en) | Attribution of congestion contributions | |
| WO2020090474A1 (en) | Packet forwarding apparatus, method and program | |
| CN113315720B (en) | Data flow control method, system and equipment | |
| Lu et al. | CHOKeR: A novel AQM algorithm with proportional bandwidth allocation and TCP protection | |
| WO2020026983A1 (en) | Packet transfer apparatus, method, and program | |
| Imputato et al. | Design and implementation of the traffic control module in ns-3 | |
| CN106899514B (en) | Queue scheduling method for guaranteeing multimedia service quality | |
| CN108989233B (en) | Congestion management method and device | |
| Zoriđ et al. | Fairness of scheduling algorithms for real-time traffic in DiffServ based networks | |
| Baker et al. | On queuing, marking, and dropping | |
| CN111092825B (en) | Method and device for transmitting message | |
| Astuti | Packet handling | |
| Domzal et al. | Approximate flow-aware networking | |
| Bodamer | A scheduling algorithm for relative delay differentiation | |
| Nandhini | Improved round robin queue management algorithm for elastic and inelastic traffic flows | |
| Domżał et al. | Efficient congestion control mechanism for flow‐aware networks | |
| CN113242606B (en) | Low-delay data transmission method and device | |
| WO2022022222A1 (en) | Method for sending data packet, and network device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |